How can I find out what theories are 'widely accepted' in the scientific world at the moment? When I read about a theory, or read a paper, how can I tell what 'the scientific world' thinks of these?
When hearing about development or history of science, the phrase 'widely accepted' is quite popular. One of the common stories seems to be of some discovery which nobody wanted to believe when published, but a decade or a few later the scientific world came to accept it and it turned into consensus. Prime examples in history would be the heliocentric world view or the theory of evolution.
More recently, I have read 'The Selfish Gene' by Richard Dawkins. He takes every chance to provide evidence that the theory of group selection is unnecessary to explain evolution. The book was published in 1976, when group selection was apparently quite popular. In annotations of the current edition, he gives comments about how the scientific world has now more accepted that group selection is wrong. I asked a question about one of his claims which I did not fully understand (irrelevant here) on Biology.SE and - unexpectedly - promptly got answers by people claiming that group selection was presently very widely accepted.
Firstly, the "official" ways that researchers state their opinion is by stating it in either a publication or in a presentation at a conference. It is from here that we can gain any idea of what any researcher thinks about a given topic.1
That being said, the only way to know whether something is "widely accepted" is to be familiar with most of the current (academic) literature on a given topic. Only by really having read the publications of most of the preeminent researchers in a given field can you really get a sense of what the consensus is on that topic. This, of course, requires a lot of time and expertise (and, if you don't have access to a university library with journal subscriptions, money). It took me years as a graduate student to learn what was the consensus about certain topics within my own field.
Because of this, anyone can state anything and most people won't know the difference. The easiest way to really determine if it's true is to ask two or three recognized researchers if they agree with the statement of interest. You can do this via email, but they likely won't get back to you. You can ask grad students in their lab and the answer will be that much less reliable, but still close to the source. Beyond that, you'll just have to find experts you can trust and rely on them.
1 Obviously, researchers are people like anyone else, and will discuss their pet theories with friends and colleagues over email and in person; it's very hard to quantify these, and you only rarely see this sort of discourse being formally recognized. Occasionally a publication will cite "unpublished discourse", but that's a rare occasion.